article

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and ingests arbitrary web pages when the subject is a URL (see "Step 1: URL
• If input starts with http:// or https://, use WebFetch to retrieve the content" and example "/article https://react.dev/..."), so it consumes untrusted third‑party web content that the agent reads and interprets.